[Zope-dev] WikiDot

Ken Manheimer klm@digicool.com
Thu, 28 Sep 2000 12:11:37 -0400 (EDT)


On Thu, 28 Sep 2000, Chris Withers wrote:

> > 4.Some people think Wiki discussions are easily dispersed. Bad Wiki
> > discussions are, but discussion products are almost always dispersed by
> > nature. On many occasions I have (already) seen people summarize and
> > structure maillist discussions into a Wiki web.
> 
> Yes, but does that mean the Wiki is good for the initial discussion?

The point you've cited seems to be about something else (why wikis can
work for discussions, not why they're good for setting them up).  My
response to your challenge is that the lack of a clearly designated
document or other artifact as the focus of a discussion can, and, i
think, usually does, lead to drift in the discussion.  In weblogs, the
document at the top anchors the discussion.  In our proposals site,
the proposals serve as (evolving) subjects of the discussion.

I don't think anyone disagrees that wiki pages as they stand are
imperfect for discussions.  I strongly feel (with lots of experience)
that mailling lists are also flawed, sorta complementarily, for
getting definite results out of discussions.  Both work with some
effort, i think many of us feel that a insightful hybrid could reap
more than the benefits of both.

> > The Wikis on the
> > dev.zope site do a bit of this with delegating discussion to a
> > discussion
> > page.
> 
> Yes, but that's only advisory, it's not enforced...

What's your point?  It seems to work, though there's possibility for
it to breakdown.  We recognize that possibility is significant, and
would like to address it - hence the WikiNG proposal, and many other
expressions of desire to fix these sorts of things.  Actually, we're
sorely chafing under lack of such fixes, but we can't afford to
neglect our other commitments (clients, including zope community!) at
the moment to do so.  And we feel that we're better off, for the time
being, with fixed points as the focus of proposal discussions, for
instance.

> > Most of these points are addressed in the proposal, but what I wanted to
> > add is the notion of the necessity of integrating the three types of
> > discussion into one product. That would make for a new generation. How
> > would that look then:
> 
> <snip Memepool equivalent> 
> 
> Funny you should say that. Some other people I'm chatting to have come
> up with roughly the same ideas...

We discussed (in a off-list discussion) a point of karls, that the
tracker and weblog can map onto eachother, with some structural
provisions.  I can see something very similar with wikis.

(This came out of a discussion with ethan - we were wondering how the
WikiNG stuff fits in with the PTK.  We came up with a bunch of stuff
that i really like, including this vision of integration with
s{qu,w}ishdot - just a concept, to demonstrate the kinds of things
that would be possible - i'm not looking to impose it on you if it
doesn't seem suitable to you.)

A weblog could constitute a wiki space, with the topic message being
the FrontPage and the threads hanging off of it being nested wiki
pages.  The discussion pages would automatically get names, per their
nesting status - LogSub1Sub2 or something, overridable by the page
author to have semantic significance.  The benefits:

 - Structured text.

 - Easy for authors to refer to other messages in the discussion by
   name, using wiki references.

 - Wiki organizational features - table-of-contents view, ability to
   adjust nesting situation in the discussion (modulo the weblogs
   policy - often the owner may disallow any such adjustment, but in
   some cases it would make sense).

 - With WikiNG's prospective notification mechanism, people could
   "subscribe" to email notifications for any changes - additions,
   edits, etc - within threads/subthreads, or just for additions at
   top levels of threads.  The latter is like "executive summary"
   monitoring, wanting to know only when the uppermost parts of the
   discussions change, without having to worry about changes to the
   outlying parts of the discussion tree.

 - With email-in wiki pages [i have to comment on mj's proposal, i
   have similar ideas in the WikiNG proposal, but haven't had time to
   evaluate his specifics!], people could use their subscriptions to
   hierarchies within the weblog to make their contributions via email 
   - often for submitting new messages, but perhaps sometimes for
   annotating existing ones, a la...

 - With WikiNG editing policy control, the weblog *could* have a
   policy that allows authors to amend their messages - eg, to insert
   or append comments to their existing messages.  Or to edit their
   messages, if that fits the task at hand.  (Eg, if what they're
   doing is collaboratively developing a document.)  The basic case
   would not allow this, for more classic, conventional weblog
   conduct.

Lots of good stuff.  The essential thing in this perspective is seeing
the weblog as a structured viewing mode for the contents of the wiki -
or the wiki as being an substrate for the contents, with enriched
content.  I imagine this would generalize to other discussion formats.

These are the kinds of things i'm hoping to get at with WikiNG - a
smart content widget, with two essential features - good impedence
matching to authoring structured, linked content (structured text plus 
wiki refs), and intrinsically determined and easily adjustable
organization.

Ken
klm@digicool.com